dolfin team mailing list archive

[Anders Logg <logg@xxxxxxxxx>]

On Tue, Aug 23, 2005 at 02:34:08PM +0200, Johan Jansson wrote:
> On Sat, Aug 20, 2005 at 05:35:37AM -0500, Anders Logg wrote:
> > I met with Johan Hoffman in Stockholm yesterday and we decided it's
> > about time we put a manual together for DOLFIN. (And the code is now
> > mature enough that it will actually be possible to do it.) I will be
> > traveling this and the next week, but we could start thinking about
> > what we want to put into the manual.
> 
> Yes, I also think it's a good idea. I'm willing to work on it in a few
> weeks.

Great. I will create a new document class fenicsmanual (based on the
old) that we could reuse for other manuals.

> > First some random thoughts:
> > 
> > 1. There is already a LaTeX-template in the source tree that I think
> > we should use.
> > 
> > 2. Manual (PDF) is automatically updated on the web page from CVS once
> > a day.
> > 
> > 3. We should probably also generate HTML from the PDF or directly from
> > tex. Is latex2html the right tool for this or are there any other
> > tools to consider?
> 
> As far as I know, latex2html is the only practical Latex to HTML
> tool. It's not always beautiful, but it produces functional output in
> my experience. I don't know about PDF to HTML tools, that would be
> nice if it worked.
> 
> I think we should put forward the PDF manual as the primary format,
> then if we can generate other formats, that's a bonus.

I agree, let's focus on the PDF.

> > 4. Should we try to separate material between the FFC and DOLFIN
> > manuals or should they overlap? We could put each section in a
> > separate tex-file and use \input{} and let some sections be identical.
> 
> I think we should keep them separate and copy relevant sections as you
> say. FFC will likely grow in complexity (added operators, output
> formats, library API etc.), and then it's going to require a manual on
> its own anyway. It's also nice to keep the Fenics component separate
> as we have done so far, which enforces good interfaces.

We probably need a short summary of the notation used by FFC in the
DOLFIN manual, but we could leave the details for the FFC manual.

/Anders

> The table of contents looks fine to me.
> 
>   Johan
> 
> _______________________________________________
> DOLFIN-dev mailing list
> DOLFIN-dev@xxxxxxxxxx
> http://www.fenics.org/cgi-bin/mailman/listinfo/dolfin-dev
> 

-- 
Anders Logg
Research Assistant Professor
Toyota Technological Institute at Chicago
http://www.tti-c.org/logg/